In contrast to a relatively rapid recovery of cells of innate immunity such as granulocytes, both the absolute levels and function of B and T lymphocytes remain abnormal for many months (Eyrich et al., 2001; Peggs and Mackinnon, 2004). T cells are known to possess some graft-versus-malignancy effect and to play an important role in the defense against viral pathogens (Bader et al., 2004; Peggs and Mackinnon, 2004). However, there is a growing body of evidence suggesting that adaptive immunity, in particular CD4⁺ T lymphocytes, has an important function in the host response against fungi. For example, non-neutropenic patients with advanced AIDS have a high risk for invasive aspergillosis (Denning, 1998). In addition, in the transplant setting, the median time of invasive aspergillosis, invasive candidiasis, and zygomycosis is 82 days (range 3–6542 days), 77 days (range 0–2219 days), and 173 days (range 7–2254 days) after HSCT, respectively, a time, when neutropenia and mucositis have generally resolved, but adaptive immune responses are still hampered (Hamza et al., 2004; Neofytos et al., 2009). An early study demonstrated that the presence of CD4⁺ T lymphocytes, which secrete interferon-gamma (IFN-γ) upon stimulation with A. fumigatus antigens indicating a T helper type 1 (T_H1) cell response, correlated with a favorable outcome (Hebart et al., 2002). It has also been shown that for up to 1 year after HSCT, transplant recipients have a significantly reduced number of functionally active anti-Aspergillus T cells compared to healthy controls, and lymphopenia is associated with a higher risk of death of transplant recipients with invasive fungal infection (Beck et al., 2008; Mikulska et al., 2009).These data provide the rationale of adoptively transferring in vitro manufactured anti-Aspergillus T cells for prevention or for treatment of Aspergillus infections. In the first clinical trial, ten haploidentical transplant recipients with evidence of invasive aspergillosis (e.g., pulmonary infiltrates, positive galactomannan test) received anti-Aspergillus T cells in the dose range of 1 × 10⁵ to 1 × 10⁶ per kg body weight between 17 and 37 days after transplantation (Perruccio et al., 2005). In all 10 patients, galactomannan positivity disappeared within 6 weeks of infusion of anti-Aspergillus T cells whereas it persisted in the 13 controls not receiving immunotherapy for as long as monitored. In addition, 9 out of the 10 patients who had received anti-Aspergillus T cells cleared invasive aspergillosis, whereas this was seen in only 7 out of the 13 controls. Anti-Aspergillus T cell clones were prepared by using heat-inactivated conidia for stimulation followed by limiting dilution which involved complex cell manipulation.

Consequently, there was much interest in improving the techniques for the clinical scale-generation of anti-Aspergillus T cells. Tramsen et al. (2009) used an Aspergillus extract for stimulation; pathogen-specific antifungal cells were selected by means of the IFN-γ secretion assay and expanded. Out of a total of 1.1 × 10⁹ white blood cells from a leukapheresis product, a median number of 2 × 10⁷ CD4⁺ T cells were obtained within 13 days. The cultured anti-Aspergillus cells exhibited almost exclusively a memory activated T_H cell phenotype. Upon re-stimulation, the generated cells produced IFN-γ, but not interleukin (IL)-4 or IL-10, indicating a T_H1 population. In addition, the cells were not end-differentiated T cells, since they proliferated upon re-stimulation. Interestingly, the functionally active anti-Aspergillus T_H1 cells elicited limited cross-reactivity with other Aspergillus species and fungi (Beck et al., 2006). As compared to unselected CD4⁺ cells, the generated anti-Aspergillus T cells exhibited reduced alloreactivity in vitro (Tramsen et al., 2009).

Recently, a simple, robust and clinically applicable procedure of generating anti-Aspergillus T cells was reported using an environmental strain of A. fumigatus and materials and reagents for clinical manufacture (Gaundar et al., 2012). In this protocol, cells were expanded with a cocktail of IL-2, IL-7, and IL-15, which resulted in a 30-fold increase in cell numbers over 21 days of culture. Generated cells were predominantly effector and central memory CD4⁺ T cells, which produced T_H1 and T_H17 cytokines and expanded upon re-stimulation.

Although most groups have employed lysates from Aspergillus isolates for the generation of protecting anti-Aspergillus T cells (Perruccio et al., 2005; Tramsen et al., 2009; Khanna et al., 2011; Gaundar et al., 2012), the use of antigen extracts may be considered problematic from a regulatory standpoint. On the other hand, whereas the immunogenic antigens of viruses such as adenovirus, cytomegalovirus (CMV) or Epstein–Barr virus (EBV) are well described, the antigenic properties of A. fumigatus are rather complex and only a few of the hundreds of (glyco)proteins of the fungus reported in the literature have been characterized at a molecular and biochemical level (Latgé, 1999). It has recently been demonstrated that different fungal components are endowed with a distinct, yet overlapping, capacity to activate protective and non-protective T_H cell responses in mice and humans (Bozza et al., 2009). In addition, an immunodominant epitope derived from A. fumigatus cell wall glucanase Crf1, restricted to three common major histocompatibility complex class II alleles, HLA-DRB1* 03, -04, and -13, has been identified which induces cross-reactive CD4⁺ T_H1 immunity not only against A. fumigatus, but also against C. albicans (Stuehler et al., 2011). However, beside the restriction to donors with certain HLA-alleles, it is important to note that the Aspergillus cell wall is a highly dynamic structure with the ability to adapt to different environments. Depending on the specific environment, each of the fungal morphotypes may exhibit different biological features, which in turn, can influence the pathogenicity of the pathogen and may result in the differential expression of antigens on its surface. Therefore, the fungal pathogen could easily escape from adoptive immunotherapy using antifungal T cells recognizing only a single antigen, whereas an approach using a fungal extract which includes multiple antigens for generating antifungal T cells might decrease this risk. To this end, the specific advantages and limitations of the use of a lysate from Aspergillus isolates and of single or multiple antigens have to be evaluated, and further studies will show which will be the most feasible approach in the clinical setting.

Unfortunately, in the clinical setting, no distinct fungal pathogen can be isolated and characterized in the majority of patients with a suspected invasive fungal infection; this, however, is considered to be a prerequisite for the determination of the antigen in order to generate pathogen-specific T cells. In addition, imaging studies such as computed tomography (CT) scan or the detection of a fungal antigen such as galactomannan, does not prove an infection due to a specific fungal pathogen. Lastly, a considerable number of patients suffer from co-infection with several species or genera of fungi. Therefore, similar to the approach in antiviral immunotherapy using T cell populations that target CMV, EBV, and adenovirus simultaneously (Hanley et al., 2009), the generation of multi-specific T cells which target a variety of different clinically important fungi such as Aspergillus spp., Candida spp., and Mucormycetes would be of major advantage. We recently reported on an approach of generating multi-specific human antifungal T cells after simultaneous stimulation with cellular extracts of A. fumigatus, C. albicans, and Rhizopus oryzae (Tramsen et al., 2013a). These generated cells consisted of activated memory T_H1 cells and reproducibly responded with IFN-γ production to a broad-spectrum of medically important fungal pathogens, such as A. fumigatus, A. niger, Penicillium chrysogenum, C. albicans, C. tropicalis, M. circinelloides, Rhizomucor pusillus, Rhizopus oryzae, Rhizopus microsporus, and Rhizopus microsporus-oligosporus. Upon re-stimulation, the generated T cells proliferated and enhanced antifungal activity of phagocytes, and showed reduced alloreactivity as compared to the original cell fraction. The ultimate goal would be to generate antifungal T cells against all medically important fungi.

Various antifungal agents have been shown to influence the function of the host immune response. For instance, early studies demonstrated that itraconazole suppresses random movement and chemotaxis of neutrophils (Vuddhakul et al., 1990), and liposomal amphotericin B and amphotericin B-deoxycholate show different immunoregulatory effects on human peripheral blood mononuclear cells (Reyes et al., 2000). In addition, liposomal amphotericin B suppresses specific activity of cytotoxic CD8⁺ T cells in the setting of murine listeriosis (Kretschmar et al., 2001). In contrast, we recently reported that various concentrations of commonly used antifungal compounds such as amphotericin B-deoxycholate, liposomal amphotericin B, fluconazole, voriconazole, posaconazole, and caspofungin did not significantly influence the secretion of IFN-γ and tumor necrosis factor-alpha (TNF-α) by human anti-Aspergillus and human anti-Candida T_H1 cells (Tramsen et al., 2013b). The proliferation of these cells was slightly decreased by posaconazole at high concentrations only, which, however, did not reach statistical significance. These data suggest that antifungal T cells can be safely administered together with commonly used antifungal compounds.

Since natural killer (NK) cells are able to kill tumor cells in vitro, there is increasing interest in using NK cells as adoptive immunotherapy against malignancies in HSCT recipients. In contrast to adoptively transferred donor-derived T cells, which are associated with the risk of GvHD, NK cells are usually well tolerated and may even mitigate GvHD (Passweg et al., 2006). In addition to the antitumor effect, NK cells exhibit cytotoxicity against virus-infected cells and activity against bacteria such as Staphylococcus aureus and against various parasites (Biron, 1997; Lieke et al., 2004; Small et al., 2008). There is also growing evidence of in vitro and animal studies that NK cells play an important role in the host response against fungal pathogens. For example, in vitro data demonstrate that NK cells are able to damage Aspergillus spp. and Rhizopus oryzae (Bouzani et al., 2011; Schmidt et al., 2011, 2012). Importantly, hyphae of both fungi are damaged by both freshly isolated and IL-2 pre-stimulated NK cells, whereas conidia are not affected (Bouzani et al., 2011; Schmidt et al., 2011, 2012). The in vitro data on the damage of Aspergillus spp. by NK cells are supported by animal studies (Morrison et al., 2003; Park et al., 2009). For example, in neutropenic mice suffering from pulmonary aspergillosis, the depletion of NK cells by antibodies resulted in a greater than twofold increase in mortality and markedly reduced clearance of the pathogen from the lungs (Morrison et al., 2003). Similarly, depletion of NK cells reduced lung IFN-γ levels and subsequently increased fungal load, whereas the transfer of activated NK cells from wild-type, but not from IFN-γ-deficient mice resulted in greater pathogen clearance from the lungs, which supports the importance of functionally active NK cells in the antifungal host response (Park et al., 2009). If further studies evaluating the effect and side effects of adoptively transferred NK cells to an immunocompromised host with invasive fungal infection will demonstrate a benefit, NK cells might become an interesting tool in immunotherapeutic antifungal strategies.